French : replace all apostrophes by a single quotation mark

[Bricolas]

How can I activate the rule to correct the apostrophe by the right single quotation mark?
In French, the apostrophe is not correct.
https://fr.wikipedia.org/wiki/Apostrophe_(typographie).

[Beurt]

I’m also very curious about the answer to this question. Those rules don’t seem to be activated, neither on my local server, nor on LibreOffice Extension:

• https://community.languagetool.org/rule/show/APOS_TYP?lang=fr&subId=1
• https://community.languagetool.org/rule/show/TYPOGRAPHICAL_APOSTROPHE?lang=en&subId=1

[danielnaber]

https://community.languagetool.org/rule/show/APOS_TYP?lang=fr&subId=1
https://community.languagetool.org/rule/show/TYPOGRAPHICAL_APOSTROPHE?lang=en&subId=1

These rules are active only in picky mode. I'm not sure if the LibreOffice add-on has an option to activate that. The browser add-on and API have.

[Beurt]

I'm sorry, but I think it is not working with the browser add-on. My browser add-on is set on picky (“Méticuleux” in French) and as you can see, the apostrophe in this very text is not curly, nor detected.

[danielnaber]

I missed that these have default="off" in their definition. I don't think they can be activated without changing the XML in the LT code, i.e. removing default="off" from these rules.

[Beurt]

Thank you for the fast answer. This is very interesting.

I noticed that it was written in those rules “By default, this rule is not active in LanguageTool”, but, I also noticed the documentation : “By default, all rules are enabled.”.

Apparently, no. All the rules are not enabled by default!

But, that leads me to another question: why?

• Why adding rules that cannot be enabled by the common user?
• Why are these rules disabled?

[Beurt]

Scanning rules/fr/style.xml I see a lot of interesting rules that are default="off". Why?

e.g.:

        <rulegroup id="ECRITURE_INCLUSIVE" name="écriture inclusive" tone_tags="general">
            <rule>
                <pattern>
                    <token postag="N.*" postag_regexp="yes">
                        <exception regexp="yes">(?-i)[A-Z].*</exception>
                        <exception postag="V.* (ind|con|sub).*" postag_regexp="yes"/></token>
                    <token regexp="yes">ou|et</token>
                    <token><match no="0"/>e
                    <exception regexp="yes">(?-i)[A-Z].*</exception>
                    <exception postag="V.* (ind|con|sub).*" postag_regexp="yes"/></token>
                </pattern>
                <message>Par politesse, le féminin précède le masculin en français.</message>
                <suggestion>\3 \2 \1</suggestion>
                <suggestion><match no="1" postag="(N) ([fm]) ([ps])" postag_regexp="yes" postag_replace="$1 m $3"/>·e</suggestion>
                <example correction="étudiante ou étudiant|étudiant·e">Pour l'<marker>étudiant ou étudiante</marker>.</example>
            </rule>
            <rule default="off">
                <antipattern>
                    <token postag="J . sp?" postag_regexp="yes"/>
                    <token>son</token>
                </antipattern>
                <pattern>
                    <token postag="D m s">
                        <exception regexp="yes">tel|telle|nul|(?-i)CE</exception></token>
                    <token regexp="yes">et|ou</token>
                    <token postag="D f s"/>
                    <token postag="N e s">
                        <exception scope="next" spacebefore="no" regexp="yes">[\[\(]</exception></token>
                </pattern>
                <message>Une autre structure peut sembler plus percutante.</message>
                <suggestion>les \4s</suggestion>
                <example correction="les photographes">Il parle avec <marker>le ou la photographe</marker>.</example>
            </rule>
            <rule>
                <pattern>
                    <token>le</token>
                    <token regexp="yes">et|ou</token>
                    <token postag="D f s"/>
                    <token postag="N [fm] s" postag_regexp="yes">
                        <exception scope="next" spacebefore="no" regexp="yes">[\[\(]</exception></token>
                </pattern>
                <message>Une autre structure peut sembler plus percutante.</message>
                <suggestion>la personne \4</suggestion>
                <suggestion>\1 <match no="4" postag="(N) ([fm]) (s)" postag_regexp="yes" postag_replace="$1 ([em]) $3"/>·e</suggestion>
                <example correction="la personne professeur|le professeur·e">Il parle avec <marker>le ou la professeur</marker>.</example>
            </rule>
            <rule>
                <pattern>
                    <token>un
                        <exception scope="previous">l'</exception></token>
                    <token regexp="yes">et|ou</token>
                    <token postag="D f s"/>
                    <token postag="N [fm] s" postag_regexp="yes">
                        <exception scope="next" spacebefore="no" regexp="yes">[\[\(]</exception></token>
                </pattern>
                <message>Une autre structure peut sembler plus percutante.</message>
                <suggestion>la personne \4</suggestion>
                <suggestion>un·e <match no="4" postag="(N) ([fm]) (s)" postag_regexp="yes" postag_replace="$1 ([em]) $3"/>·e</suggestion>
                <example correction="la personne professeur|un·e professeur·e">Il parle avec <marker>un ou une professeur</marker>.</example>
            </rule>
        </rulegroup>

or

        <rule id="MAIS_SENT_START_2" name="'Mais' en début de phrase" tone_tags="formal" default="off">
            <antipattern>
                <token postag="SENT_START"/>
                <token skip="3">mais</token>
                <token postag="SENT_END"/>
            </antipattern>
            <antipattern>
                <token postag="SENT_START"/>
                <token>mais</token>
                <token postag="A"><exception regexp="yes">beaucoup|contrairement|déjà|devant|derrière</exception></token>
                <token postag="[DPZR].*" postag_regexp="yes"/>
            </antipattern>
            <antipattern>
                <token postag="SENT_START"/>
                <token>mais</token>
                <token>bon</token>
                <token postag="[DPZR].*" postag_regexp="yes"/>
            </antipattern>
            <pattern>
                <token postag="SENT_START"/>
                <marker>
                    <token case_sensitive="yes">Mais</token>
                    <token><exception negate="yes" regexp="yes">\p{L}[\p{L}'-]*</exception></token>
                </marker>
            </pattern>
            <message>L'utilisation de "Mais" en première position dans une phrase n'est pas incorrecte, mais plutôt un parti pris stylistique. Vous pouvez l'accompagner d'une virgule pour marquer une pause.</message>
            <suggestion>\3</suggestion>
            <suggestion>\2, \3</suggestion>
            <url>https://www.academie-francaise.fr/car-en-debut-de-phrase#:~:text=Les%20conjonctions%20de%20coordination%20ne,pas%20de%20m%C3%AAme%20pour%20car.</url>
            <example correction="Il|Mais, il"><marker>Mais il</marker> lui était impossible de revenir en arrière.</example>
            <example correction="Çağan|Mais, Çağan"><marker>Mais Çağan</marker> était carrément vert.</example>
            <example>Il me semble que cet homme est inconscient, mais il n'a pas l'air d'être blessé.</example>
            <example>Mais peu importe.</example>
            <example>Mais bon Sophie est rentré à nouveau chez elle.</example>
        </rule>

There is also a lot in grammar.xml.

[Beurt]

In a very old documentation (https://web.archive.org/web/20120102085443/http://www.languagetool.org/development/) I found this:

4. Turning the rule off

Some rules can be optional, useful only in specific registers, or very sensitive. You can turn them off by default by using an attribute default="off". The user can turn the rule in the Options dialog box, and this setting is being saved in the configuration file.

It does not seem to be accurate anymore.

And I still don't understand why there is the default="off" attribute (I understand the `default="temp_off" attribute!).

[LucieSteib]

Hi :)
The apostrophe rule was deactivated as, it was pointed before, apostrophes are extremely frequent in French.
The rule for apostrophes is off for two main reasons:

• We don't have yet a way to "turn all apostrophes" to the correct typographical one. It was extremely annoying to click on every apostrophe to correct.
• The error counter was displaying hundreds of errors that were all apostrophes.
  About the defaut="off", it's to avoid seeing the rule in the nightly regression test, that doesn't change much outside of this.
  Hope this helps!

[Beurt]

Thanks for the answer!

Too bad there is no GUI to switch these interesting rules (that are existing) back on.